Chest X-Ray (WIP)

Purpose: To create usable Chest Xrays with a GAN as a way to augment medical data without needing to worry about privacy and accessibility.

How:

1. Create an implementation of DenseNet in PyTorch from : https://arxiv.org/abs/1608.06993
2. Create an implementation of StyleGan in PyTorch from : https://github.com/NVlabs/stylegan
3. Generate images from StyleGan
4. Train the DenseNet two different times, with the real images and the StyleGan generated images
5. Compare the scores

What data is being used:

All the data was used can be found here: https://www.kaggle.com/nih-chest-xrays/data

Technologies: PyTorch, StyleGAN, DenseNet

Tasks:

Task	Complete
Gather Data	✔️
Code DenseNet	✔️
Code StyleGan
Generate images
Train two DenseNets
Analyze results